The investigators state that the rapid lethal infections and malignancies in patients with HIV infection prompts the use of many varied drugs and drug combinations. Many of these drugs have not been fully evaluated and few have been evaluated in the multiple combinations that occur in clinical practice. This application describes a data-based approach to the investigation of the clinical pharmacology of these drugs and their combinations. The general approach will be to conduct pilot studies to screen for drug interactions or predictors of adverse reactions. Clinical pharmacology studies will then be conducted using specifically designed protocols in larger numbers of patients. The specific aims are as follows: (1) The first is to detect and quantify pharmacokinetic and dynamic interactions between drugs used to treat patients with HIV infection. (2) Second, the applicants will evaluate the pharmacokinetics of these drugs in special populations of patients, including minorities, women, and those with malnutrition, alcoholism, intravenous drug use, AIDS gastroenteropathy and AIDS cardiomyopathy. (3) Next, the applicants will attempt to identify sources of inter-individual variation in response and adherence to therapy. Since genetic polymorphisms are responsible for adverse or aberrant responses to a large number of drugs, all patients enrolled at this ACTG site will be asked to volunteer for simple tests with safe prototypic drugs to identify their phenotype for polymorphic drug metabolism (e.g., N-acetylation, CYP2D6, CYP3A4, mephenytoin P450). For patients who cannot be drug free during baseline testing, their CYP2D6 genotype will be determined from a 0.5-ml blood sample using a polymerase chain reaction (PCR)-based assay. The predictive power of each of these polymorphisms will be examined by comparing their usual distributions to those in patients who develop adverse drug reactions (ADRs). The results of these studies, the applicants state, should make possible the identification of individuals at increased risk of ADRs or drug interactions. Additionally, the biological factors will be analyzed to test the hypothesis that they may influence the behavior of individuals in deciding to receive investigational therapy or to remain in, or comply with, a study protocol. Although the exact studies to be performed can only be determined after consultation and review by the ACTU coordinating committees, the applicants propose herein a specific study to evaluate the relationship between foscarnet (PFA), serum electrolyte changes and cardiac electrophysiology. Also general protocols for the evaluation of drug-drug interactions (specific aim #1) and pharmacokinetics in special populations (specific Aim #2) are described. The information from these studies should allow physicians caring for patients with HIV infection to provide therapy based on a greater understanding of its actions and risks for adverse effects.